shreyajn commited on
Commit
3108d1c
·
verified ·
1 Parent(s): 16ffcac

Upload README.md with huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +77 -105
README.md CHANGED
@@ -36,64 +36,64 @@ More details on model performance across various devices, can be found
36
 
37
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
- | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.135 ms | 0 - 30 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
40
- | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.03 ms | 2 - 4 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.so) |
41
- | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.578 ms | 0 - 183 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
42
- | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.447 ms | 0 - 129 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
43
- | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.497 ms | 0 - 21 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.so) |
44
- | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.867 ms | 0 - 134 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
45
- | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.216 ms | 0 - 61 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
46
- | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.237 ms | 3 - 64 MB | FP16 | NPU | Use Export Script |
47
- | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.755 ms | 3 - 69 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
48
- | TrOCRDecoder | SA7255P ADP | SA7255P | TFLITE | 11.828 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
49
- | TrOCRDecoder | SA7255P ADP | SA7255P | QNN | 11.69 ms | 7 - 16 MB | FP16 | NPU | Use Export Script |
50
- | TrOCRDecoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.097 ms | 0 - 52 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
51
- | TrOCRDecoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.023 ms | 2 - 4 MB | FP16 | NPU | Use Export Script |
52
- | TrOCRDecoder | SA8295P ADP | SA8295P | TFLITE | 2.973 ms | 0 - 52 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
53
- | TrOCRDecoder | SA8295P ADP | SA8295P | QNN | 2.919 ms | 7 - 25 MB | FP16 | NPU | Use Export Script |
54
- | TrOCRDecoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.086 ms | 0 - 31 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
55
- | TrOCRDecoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.038 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
56
- | TrOCRDecoder | SA8775P ADP | SA8775P | TFLITE | 3.12 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
57
- | TrOCRDecoder | SA8775P ADP | SA8775P | QNN | 3.009 ms | 6 - 17 MB | FP16 | NPU | Use Export Script |
58
- | TrOCRDecoder | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 11.828 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
59
- | TrOCRDecoder | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 11.69 ms | 7 - 16 MB | FP16 | NPU | Use Export Script |
60
- | TrOCRDecoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.068 ms | 0 - 85 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
61
- | TrOCRDecoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.02 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
62
- | TrOCRDecoder | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 3.12 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
63
- | TrOCRDecoder | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 3.009 ms | 6 - 17 MB | FP16 | NPU | Use Export Script |
64
- | TrOCRDecoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 2.775 ms | 0 - 123 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
65
- | TrOCRDecoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 2.784 ms | 5 - 130 MB | FP16 | NPU | Use Export Script |
66
- | TrOCRDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.205 ms | 7 - 7 MB | FP16 | NPU | Use Export Script |
67
- | TrOCRDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.331 ms | 68 - 68 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
68
- | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 38.902 ms | 7 - 31 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
69
- | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 36.585 ms | 2 - 4 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.so) |
70
- | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 38.574 ms | 14 - 123 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
71
- | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 29.975 ms | 5 - 171 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
72
- | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 29.869 ms | 2 - 22 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.so) |
73
- | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 30.367 ms | 14 - 127 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
74
- | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 27.693 ms | 6 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
75
- | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 26.389 ms | 2 - 165 MB | FP16 | NPU | Use Export Script |
76
- | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 26.791 ms | 14 - 123 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
77
- | TrOCREncoder | SA7255P ADP | SA7255P | TFLITE | 256.442 ms | 6 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
78
- | TrOCREncoder | SA7255P ADP | SA7255P | QNN | 248.286 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
79
- | TrOCREncoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 39.664 ms | 7 - 30 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
80
- | TrOCREncoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 37.716 ms | 2 - 4 MB | FP16 | NPU | Use Export Script |
81
- | TrOCREncoder | SA8295P ADP | SA8295P | TFLITE | 53.193 ms | 0 - 162 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
82
- | TrOCREncoder | SA8295P ADP | SA8295P | QNN | 50.2 ms | 2 - 20 MB | FP16 | NPU | Use Export Script |
83
- | TrOCREncoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 39.12 ms | 7 - 27 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
84
- | TrOCREncoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 36.737 ms | 2 - 4 MB | FP16 | NPU | Use Export Script |
85
- | TrOCREncoder | SA8775P ADP | SA8775P | TFLITE | 47.664 ms | 7 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
86
- | TrOCREncoder | SA8775P ADP | SA8775P | QNN | 42.585 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
87
- | TrOCREncoder | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 256.442 ms | 6 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
88
- | TrOCREncoder | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 248.286 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
89
- | TrOCREncoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 38.481 ms | 7 - 35 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
90
- | TrOCREncoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 36.68 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
91
- | TrOCREncoder | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 47.664 ms | 7 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
92
- | TrOCREncoder | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 42.585 ms | 2 - 12 MB | FP16 | NPU | Use Export Script |
93
- | TrOCREncoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 51.659 ms | 7 - 172 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
94
- | TrOCREncoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 51.994 ms | 2 - 170 MB | FP16 | NPU | Use Export Script |
95
- | TrOCREncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 34.406 ms | 2 - 2 MB | FP16 | NPU | Use Export Script |
96
- | TrOCREncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 37.296 ms | 51 - 51 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
97
 
98
 
99
 
@@ -158,7 +158,7 @@ TrOCRDecoder
158
  Device : Samsung Galaxy S23 (13)
159
  Runtime : TFLITE
160
  Estimated inference time (ms) : 2.1
161
- Estimated peak memory usage (MB): [0, 30]
162
  Total # Ops : 399
163
  Compute Unit(s) : NPU (399 ops)
164
 
@@ -167,7 +167,7 @@ TrOCREncoder
167
  Device : Samsung Galaxy S23 (13)
168
  Runtime : TFLITE
169
  Estimated inference time (ms) : 38.9
170
- Estimated peak memory usage (MB): [7, 31]
171
  Total # Ops : 591
172
  Compute Unit(s) : NPU (591 ops)
173
  ```
@@ -191,43 +191,26 @@ import qai_hub as hub
191
  from qai_hub_models.models.trocr import Model
192
 
193
  # Load the model
194
- model = Model.from_pretrained()
195
- decoder_model = model.decoder
196
- encoder_model = model.encoder
197
 
198
  # Device
199
- device = hub.Device("Samsung Galaxy S23")
200
 
201
  # Trace model
202
- decoder_input_shape = decoder_model.get_input_spec()
203
- decoder_sample_inputs = decoder_model.sample_inputs()
204
 
205
- traced_decoder_model = torch.jit.trace(decoder_model, [torch.tensor(data[0]) for _, data in decoder_sample_inputs.items()])
206
 
207
  # Compile model on a specific device
208
- decoder_compile_job = hub.submit_compile_job(
209
- model=traced_decoder_model ,
210
  device=device,
211
- input_specs=decoder_model.get_input_spec(),
212
  )
213
 
214
  # Get target model to run on-device
215
- decoder_target_model = decoder_compile_job.get_target_model()
216
- # Trace model
217
- encoder_input_shape = encoder_model.get_input_spec()
218
- encoder_sample_inputs = encoder_model.sample_inputs()
219
-
220
- traced_encoder_model = torch.jit.trace(encoder_model, [torch.tensor(data[0]) for _, data in encoder_sample_inputs.items()])
221
-
222
- # Compile model on a specific device
223
- encoder_compile_job = hub.submit_compile_job(
224
- model=traced_encoder_model ,
225
- device=device,
226
- input_specs=encoder_model.get_input_spec(),
227
- )
228
-
229
- # Get target model to run on-device
230
- encoder_target_model = encoder_compile_job.get_target_model()
231
 
232
  ```
233
 
@@ -239,15 +222,11 @@ After compiling models from step 1. Models can be profiled model on-device using
239
  provisioned in the cloud. Once the job is submitted, you can navigate to a
240
  provided job URL to view a variety of on-device performance metrics.
241
  ```python
242
- decoder_profile_job = hub.submit_profile_job(
243
- model=decoder_target_model,
244
- device=device,
245
- )
246
- encoder_profile_job = hub.submit_profile_job(
247
- model=encoder_target_model,
248
  device=device,
249
  )
250
-
251
  ```
252
 
253
  Step 3: **Verify on-device accuracy**
@@ -255,20 +234,13 @@ Step 3: **Verify on-device accuracy**
255
  To verify the accuracy of the model on-device, you can run on-device inference
256
  on sample input data on the same cloud hosted device.
257
  ```python
258
- decoder_input_data = decoder_model.sample_inputs()
259
- decoder_inference_job = hub.submit_inference_job(
260
- model=decoder_target_model,
261
- device=device,
262
- inputs=decoder_input_data,
263
- )
264
- decoder_inference_job.download_output_data()
265
- encoder_input_data = encoder_model.sample_inputs()
266
- encoder_inference_job = hub.submit_inference_job(
267
- model=encoder_target_model,
268
  device=device,
269
- inputs=encoder_input_data,
270
  )
271
- encoder_inference_job.download_output_data()
272
 
273
  ```
274
  With the output of the model, you can compute like PSNR, relative errors or
 
36
 
37
  | Model | Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model
38
  |---|---|---|---|---|---|---|---|---|
39
+ | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 2.061 ms | 0 - 27 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
40
+ | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 2.083 ms | 2 - 4 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.so) |
41
+ | TrOCRDecoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 2.55 ms | 0 - 128 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
42
+ | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 1.512 ms | 0 - 135 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
43
+ | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 1.509 ms | 0 - 18 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.so) |
44
+ | TrOCRDecoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 1.766 ms | 0 - 135 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
45
+ | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 1.228 ms | 0 - 61 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
46
+ | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 1.404 ms | 3 - 65 MB | FP16 | NPU | Use Export Script |
47
+ | TrOCRDecoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 1.678 ms | 3 - 152 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
48
+ | TrOCRDecoder | SA7255P ADP | SA7255P | TFLITE | 11.747 ms | 0 - 60 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
49
+ | TrOCRDecoder | SA7255P ADP | SA7255P | QNN | 11.72 ms | 7 - 16 MB | FP16 | NPU | Use Export Script |
50
+ | TrOCRDecoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 2.083 ms | 0 - 30 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
51
+ | TrOCRDecoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 2.041 ms | 1 - 3 MB | FP16 | NPU | Use Export Script |
52
+ | TrOCRDecoder | SA8295P ADP | SA8295P | TFLITE | 3.033 ms | 0 - 52 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
53
+ | TrOCRDecoder | SA8295P ADP | SA8295P | QNN | 2.893 ms | 7 - 25 MB | FP16 | NPU | Use Export Script |
54
+ | TrOCRDecoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 2.1 ms | 0 - 30 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
55
+ | TrOCRDecoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 2.015 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
56
+ | TrOCRDecoder | SA8775P ADP | SA8775P | TFLITE | 3.149 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
57
+ | TrOCRDecoder | SA8775P ADP | SA8775P | QNN | 3.009 ms | 7 - 17 MB | FP16 | NPU | Use Export Script |
58
+ | TrOCRDecoder | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 11.747 ms | 0 - 60 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
59
+ | TrOCRDecoder | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 11.72 ms | 7 - 16 MB | FP16 | NPU | Use Export Script |
60
+ | TrOCRDecoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 2.066 ms | 0 - 31 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
61
+ | TrOCRDecoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 2.03 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
62
+ | TrOCRDecoder | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 3.149 ms | 0 - 59 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
63
+ | TrOCRDecoder | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 3.009 ms | 7 - 17 MB | FP16 | NPU | Use Export Script |
64
+ | TrOCRDecoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 2.885 ms | 0 - 122 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.tflite) |
65
+ | TrOCRDecoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 2.985 ms | 7 - 133 MB | FP16 | NPU | Use Export Script |
66
+ | TrOCRDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 2.234 ms | 7 - 7 MB | FP16 | NPU | Use Export Script |
67
+ | TrOCRDecoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 2.322 ms | 68 - 68 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCRDecoder.onnx) |
68
+ | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | TFLITE | 38.875 ms | 7 - 29 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
69
+ | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | QNN | 36.998 ms | 0 - 2 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.so) |
70
+ | TrOCREncoder | Samsung Galaxy S23 | Snapdragon® 8 Gen 2 | ONNX | 36.962 ms | 7 - 134 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
71
+ | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | TFLITE | 29.99 ms | 6 - 174 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
72
+ | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | QNN | 30.188 ms | 2 - 20 MB | FP16 | NPU | [TrOCR.so](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.so) |
73
+ | TrOCREncoder | Samsung Galaxy S24 | Snapdragon® 8 Gen 3 | ONNX | 30.736 ms | 14 - 165 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
74
+ | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | TFLITE | 27.772 ms | 5 - 168 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
75
+ | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | QNN | 26.54 ms | 2 - 166 MB | FP16 | NPU | Use Export Script |
76
+ | TrOCREncoder | Snapdragon 8 Elite QRD | Snapdragon® 8 Elite | ONNX | 28.836 ms | 14 - 172 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
77
+ | TrOCREncoder | SA7255P ADP | SA7255P | TFLITE | 256.881 ms | 7 - 170 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
78
+ | TrOCREncoder | SA7255P ADP | SA7255P | QNN | 248.315 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
79
+ | TrOCREncoder | SA8255 (Proxy) | SA8255P Proxy | TFLITE | 38.924 ms | 8 - 32 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
80
+ | TrOCREncoder | SA8255 (Proxy) | SA8255P Proxy | QNN | 36.621 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
81
+ | TrOCREncoder | SA8295P ADP | SA8295P | TFLITE | 53.159 ms | 7 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
82
+ | TrOCREncoder | SA8295P ADP | SA8295P | QNN | 50.227 ms | 2 - 18 MB | FP16 | NPU | Use Export Script |
83
+ | TrOCREncoder | SA8650 (Proxy) | SA8650P Proxy | TFLITE | 38.668 ms | 7 - 36 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
84
+ | TrOCREncoder | SA8650 (Proxy) | SA8650P Proxy | QNN | 36.86 ms | 2 - 4 MB | FP16 | NPU | Use Export Script |
85
+ | TrOCREncoder | SA8775P ADP | SA8775P | TFLITE | 47.569 ms | 7 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
86
+ | TrOCREncoder | SA8775P ADP | SA8775P | QNN | 42.56 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
87
+ | TrOCREncoder | QCS8275 (Proxy) | QCS8275 Proxy | TFLITE | 256.881 ms | 7 - 170 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
88
+ | TrOCREncoder | QCS8275 (Proxy) | QCS8275 Proxy | QNN | 248.315 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
89
+ | TrOCREncoder | QCS8550 (Proxy) | QCS8550 Proxy | TFLITE | 38.449 ms | 7 - 32 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
90
+ | TrOCREncoder | QCS8550 (Proxy) | QCS8550 Proxy | QNN | 37.026 ms | 2 - 5 MB | FP16 | NPU | Use Export Script |
91
+ | TrOCREncoder | QCS9075 (Proxy) | QCS9075 Proxy | TFLITE | 47.569 ms | 7 - 169 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
92
+ | TrOCREncoder | QCS9075 (Proxy) | QCS9075 Proxy | QNN | 42.56 ms | 2 - 11 MB | FP16 | NPU | Use Export Script |
93
+ | TrOCREncoder | QCS8450 (Proxy) | QCS8450 Proxy | TFLITE | 48.787 ms | 7 - 171 MB | FP16 | NPU | [TrOCR.tflite](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.tflite) |
94
+ | TrOCREncoder | QCS8450 (Proxy) | QCS8450 Proxy | QNN | 50.666 ms | 2 - 168 MB | FP16 | NPU | Use Export Script |
95
+ | TrOCREncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | QNN | 34.302 ms | 2 - 2 MB | FP16 | NPU | Use Export Script |
96
+ | TrOCREncoder | Snapdragon X Elite CRD | Snapdragon® X Elite | ONNX | 36.636 ms | 50 - 50 MB | FP16 | NPU | [TrOCR.onnx](https://huggingface.co/qualcomm/TrOCR/blob/main/TrOCREncoder.onnx) |
97
 
98
 
99
 
 
158
  Device : Samsung Galaxy S23 (13)
159
  Runtime : TFLITE
160
  Estimated inference time (ms) : 2.1
161
+ Estimated peak memory usage (MB): [0, 27]
162
  Total # Ops : 399
163
  Compute Unit(s) : NPU (399 ops)
164
 
 
167
  Device : Samsung Galaxy S23 (13)
168
  Runtime : TFLITE
169
  Estimated inference time (ms) : 38.9
170
+ Estimated peak memory usage (MB): [7, 29]
171
  Total # Ops : 591
172
  Compute Unit(s) : NPU (591 ops)
173
  ```
 
191
  from qai_hub_models.models.trocr import Model
192
 
193
  # Load the model
194
+ torch_model = Model.from_pretrained()
 
 
195
 
196
  # Device
197
+ device = hub.Device("Samsung Galaxy S24")
198
 
199
  # Trace model
200
+ input_shape = torch_model.get_input_spec()
201
+ sample_inputs = torch_model.sample_inputs()
202
 
203
+ pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])
204
 
205
  # Compile model on a specific device
206
+ compile_job = hub.submit_compile_job(
207
+ model=pt_model,
208
  device=device,
209
+ input_specs=torch_model.get_input_spec(),
210
  )
211
 
212
  # Get target model to run on-device
213
+ target_model = compile_job.get_target_model()
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
214
 
215
  ```
216
 
 
222
  provisioned in the cloud. Once the job is submitted, you can navigate to a
223
  provided job URL to view a variety of on-device performance metrics.
224
  ```python
225
+ profile_job = hub.submit_profile_job(
226
+ model=target_model,
 
 
 
 
227
  device=device,
228
  )
229
+
230
  ```
231
 
232
  Step 3: **Verify on-device accuracy**
 
234
  To verify the accuracy of the model on-device, you can run on-device inference
235
  on sample input data on the same cloud hosted device.
236
  ```python
237
+ input_data = torch_model.sample_inputs()
238
+ inference_job = hub.submit_inference_job(
239
+ model=target_model,
 
 
 
 
 
 
 
240
  device=device,
241
+ inputs=input_data,
242
  )
243
+ on_device_output = inference_job.download_output_data()
244
 
245
  ```
246
  With the output of the model, you can compute like PSNR, relative errors or